Geometric distribution

[Brian Gough]

Giulio Bottazzi writes:
 > I've a simple question: why the geometric distribution, inside GSL, is
 > defined according to
 > prob(k) =  p (1 - p)^(k-1) for k = 1, 2, 3, ...
 > rather than the more canonical
 > prob(k) = p (1-p)^k for k = 0, 1, 2, ...
 > is there some deep reason I'm missing?

I don't think so, it's just the convention from Knuth.

-- 
Brian Gough

Network Theory Ltd,
Commercial support for GSL --- http://www.network-theory.co.uk/gsl/